SEARCH

SEARCH BY CITATION

Keywords:

  • bone marrow haematopoietic stem cells;
  • HGF ;
  • HSC ;
  • liver regeneration;
  • liver resection;
  • SDF-1

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Background

The molecular mechanisms of haematopoietic stem cells (HSC) mobilization and homing to the liver after partial hepatectomy (PH) remain largely unexplored.

Methods

Functional liver volume loss and regain was determined by computerized tomography (CT) volumetry in 30 patients following PH. Peripheral HSC mobilization was investigated by fluorescence-activated cell sorting (FACS) analyses and cytokine enzyme-linked immunosorbent assay assays. Migration of purified HSC towards hepatic growth factor (HGF) and stroma-derived factor-1 (SDF-1) gradients was tested in vitro. Mice after 70% PH were examined for HSC mobilization by FACS and cytokine mRNA expression in the liver. FACS-sorted HSC were administered after PH and hepatocyte proliferation was evaluated by immunohistochemical staining for Ki67.

Results

Impaired liver function was noted after extended hepatic resection when compared to smaller resections. Patients with large liver resections were characterized by significantly higher levels of peripheral HSC which were positively correlated with the extent of resected liver volume and its regain after 3 weeks. Increased plasma levels of HGF, SDF-1 and insulin like growth factor (IGF-1) were evident within the first 6 hours post resection. Migration assays of human HSC in vitro showed a specific target-demonstrated migration towards recombinant HGF and SDF-1 gradients in a concentration and specific receptor (c-Met and CXCR4) dependent manner. The evaluation of peripheral human alpha foetoprotein expression demonstrated pronounced stemness following increased CD133+HSC in the course of liver regeneration following PH. Our human data were further validated in a murine model of PH and furthermore demonstrated increased hepatocyte proliferation subsequent to CD133+HSC treatment.

Conclusion

HGF and SDF-1 are required for effective HSC mobilization and homing to the liver after hepatic resection. These findings have significant implications for potential therapeutic strategies targeting chemotactant modulation and stem cell mobilization for liver protection and regeneration.

Abbreviations
AFP

alpha-1-foetoprotein

BM

bone marrow

BMSC

bone marrow stem cell

CRP

C-reactive protein

CT

computerized tomography

d

days

ECLIA

electro chemiluminescence assay

ELISA

enzyme-linked immunosorbent assay

FACS

fluorescence-activated cell sorting

HGF

hepatic growth factor

h

hours

HSC

haematopoetic stem cells

IGF-1

insulin-like growth factor

ICG

indocyanine green

INR

international normalized ratio

MACS

magnetic-activated cell sorting

PDR

plasma disappearing rate

PH

partial hepatectomy

POD

post-operative day

SCF

stem cell factor

SDF-1

stroma-derived factor-1

TLV

total liver volume

WBC

white blood cell

Liver regeneration is an organized response of the liver to injury and involves molecular changes in morphologic structure, gene expression as well as growth factor and cytokine production [1-3].

Recent reports suggest that bone marrow and haematopoietic stem cells (HSC) participate in hepatic proliferation and liver regeneration [4-15]. To date, the specific cell types and relevant molecular mechanisms of stem cell mobilization and homing to the liver are poorly understood [16]. However, bone marrow (BM)-derived circulating cells have been experimentally shown to participate in hepatic proliferation after liver injury [8, 10, 12, 14, 15, 17]. Our own clinical experience demonstrated a therapeutic potential of administered CD133+ bone marrow stem cells (BMSC) to augment liver regeneration in the course of extended partial hepatectomy (PH) [5-7].

It is well established that the damaged liver releases chemokines and possible chemotactants, such as stroma-derived factor-1 (SDF-1), stem cell factor (SCF), hepatocellular growth factor (HGF), insulin-like growth factor (IGF-1) and others, to participate in the concert of inflammatory cell homing from extrahepatic sources to the liver in rodents and humans [4, 18-20]. HGF seems to promote murine mobilization and SCF-dependent recruitment of haematopoietic CD34+ cells from BM into the liver [21]. Elevated HGF serum levels were demonstrated after hepatectomy in healthy donors in the context of living-related liver donation [22] . It was shown that the injured liver in rodents does express increased levels of SDF-1, attracting CD133+ BMSC, that are positive for the SDF-1 receptor CXCR4 [23]. Hepatic engraftment from extrahepatic progenitor cells is accelerated in cases of experimental liver damage if contrasted with non-injured liver tissue [13]. In contrast to CD34+ HSC, it is still inconclusive whether CD133+ HSC are peripherally mobilized and locally involved in the course of human liver regeneration subsequent to hepatic resection. This may be because of small numbers of patients investigated, the lack of extended forms of resection, mixture of liver-diseased and non-diseased patients as well as a missing determination of the exact volume of resected liver tissue in these reports [24-27].

To further investigate the molecular mechanism of CD133+ HSC and mobilizing stimuli for liver regeneration following partial hepatectomy, we prospectively correlated CD133+/CD45+ HSC mobilization subsequent to human liver resection with the exact loss and regain of liver volume, determined by computerized tomography (CT) scan volumetry. Stem cell mobilization was correlated with the functional resection volume and chemotactants such as SDF-1, IGF-1, SCF and HGF that were expressed during the early course post-liver resection. Our data clearly show that the chemotactants SDF-1 and HGF specifically mobilize HSC by testing their dependency on specific antagonists CXCR4 and c-Met respectively. Alpha-1-foetoprotein (AFP) expression, as a marker of liver progenitor cell activation and stemness [28], was positively correlated with the extent of liver tissue loss following hepatectomy and preceding peripheral CD133+ cell mobilization respectively. Our findings from experiments performed on human specimens were further tested in a murine model of 70% PH. Mice after PH also demonstrated increased levels of HSC, peripherally and in the BM accompanied by increased cytokines in the liver. Increased hepatocyte proliferation was noted after therapeutical administration of HSC in vivo.

These findings might have implications for potential therapeutic strategies targeting chemotactant modulation and HSC mobilization for liver protection and regeneration.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Human studies

Patients

Thirty patients (♀ = 11/♂ = 19; median age 65 ± 25 years) undergoing hepatic resections for liver malignancies at the Department of Surgery, University Hospital of Düsseldorf, Germany were included in this study. Patients were divided into two groups depending on the extent of liver resection: group I: small hepatic resection <20% of total liver volume (TLV) (n = 17) and group II: large hepatic resection >30% of TLV (n = 13). The resection rate for large hepatic resections (group II) ranged from 34 to 65% resected liver tissue.

Written informed consent was obtained from all patients. The study was approved by the Ethics committee of the Heinrich Heine University of Düsseldorf (IRBs #2852, #2853, #2916).

Characterization and clinical course of patients were determined by surgical parameters and routine medical laboratory diagnostics regarding hepatic function (bilirubin, albumin, transaminases), coagulation [International normalized ratio (INR), fibrinogen, Quick], blood count (haemoglobin, haematocrit) and inflammation (leucocytes, CRP) values.

Hepatic CT volumetry

Data sets for liver volumetry were obtained from helical CT of the upper abdomen both prior to and 24 h, 7 d and 21 d after hepatectomy [7]. TLV (calculated hepatic volume minus tumour volume) was calculated based on these data sets, obtained in the portal vein enhancement phase.

Indocyanine green metabolism

For liver function analysis, plasma disappearance rate of indocyanine green (ICG) was measured using a LIMON Module (PULSION Medical Systems, Feldkirchen, Germany), according to manufacturer's instructions.

Fluorescence-activated cell sorting

Citrate blood was harvested prior to and 24 h, 2 d, 4 d, 7 d, 10 d and 21 d after liver resection. Whole blood (100 μl) was incubated with antibodies for 15 min, lysed (10 min; Lysing solution, BD Bioscience, San Jose, CA, USA) and washed twice (10 min, 1500 g). Cells were resuspended in PBS and CD133+/CD45+ BMSCs were analysed on a fluorescence-activated cell sorting (FACS) Canto Flow Cytometer (BD Bioscience). Following antibodies were used: anti-CD133-PE (clone: 293C3; Miltenyi Biotec, Bergisch Gladbach, Germany): 10 μl, anti-CD45-APC (clone: 2D 1, BD Bioscience): 2.5 μl, anti-CD34-PE-Cy7 (clone: 581, BD Bioscience): 5 μl and anti-CD39-FITC (clone: BU61, BD Bioscience): 2 μl.

Enzyme-linked immunosorbent assay

Patient blood was centrifuged in serum-separating tubes without anticoagulants (10 min, 1500 g/3000 rpm) and serum was stored at −80°C. Serum levels of HGF, SDF-1, IGF and SCF were determined utilizing commercially available enzyme-linked immunosorbent assays (ELISA) according to manufacturer's instructions (R&D Systems, Minneapolis, MN, USA). Samples were run in duplicates.

Electro chemiluminescence assay

Serum levels of total AFP were determined applying electro chemiluminescence assay (ECLIA) and immunological ligand assays by the central institute for clinical chemistry and laboratory diagnostics of the University Hospital Düsseldorf. Samples were run in duplicates.

Magnetic-activated cell sorting

CD133+ HSC were separated from the human bone marrow by using magnetic-activated cell sorting (MACS) (Miltenyi Biotec), according to the manufacturer's instructions. Purity was assessed by FACS to be 83.6 ± 3.9%.

Chemotaxis assays

Transwell Boyden Chamber chemotaxis assays were performed using 24-well Transwell Permeable Supports with 5.0 μm pore size (Corning Costar, Lowell, MA, USA). HSC 5 × 104cells/100 μl DME Medium + 10% (FCS; Sigma, Saint Louis, MO, USA) were added into the upper chamber. After 24 h incubation at 37°C, HSC that had transmigrated to the lower chamber containing targets (50 ng/ml HGF or 125 ng/ml SDF-1) were harvested, counted and determined by FACS analysis.

Specific, target-directed migration towards recombinant HGF/SDF-1 gradients was analysed by receptor blocking assays using the specific HGF receptor c-Met antibody SU 11274 (Sigma; 5–10 μM) and the specific SDF-1 receptor CXCR4 antagonist AMD3100 (Sigma; 5 μg/ml).

Animal Studies

Animals

Wild-type C57Bl/6 mice (Taconic, Germantown, NY, USA) aged 7–9 weeks were used in accordance with the guidelines from the American Association for Laboratory Animal Care. The Beth Israel Deaconess Medical Center Institutional Animal Care and Use Committees approved all animal research protocols.

Liver injury model

Seventy percent partial hepatectomy and sham surgery were performed as previously described [29]. Sham-operated mice served as controls.

Gene expression analyses

Total RNAs were isolated from liver tissue, reverse-transcribed, followed by quantitative PCR amplification performed in duplicates, as previously described [29]. Data were normalized to GAPH. The primer sequences used are mouse:

HGF:F-CTCCCTTCCCTACTCGGATAR-AGCAGACTGATCCCTAAAGC
SDF-1:F-GAGAAAGCTTTAAACAAGGGGCR-AAGAGGGAGGAGCGAGTTAC
IGF:F-TGCTCTTCAGTTCGTGTGR-ACATCTCCAGTCTCCTCAG
AFP:F-CACAGAAGAGGGTCCAAAGTR- GCTCACACCAAAGAGTCAAC
GAPDH:F-GACGGCCGCATCTTCTTGTR-CACACCGACCTTCACCATTTT
Isolation of bone marrow mononuclear cells

Bone marrow was flushed out from the hind legs with RMPI 1640 medium and filtered through BD Falcon cell strainers (70 μM; BD Bioscience). BM was layered over histopaque-1083 (Sigma) and centrifuged (30 min, 500 rpm, 4°C). The buffy coat was washed twice (10 min, 500 rpm, 4°C) and processed as indicated.

Fluorescence-activated cell sorting of CD133+/CD45+ HSC

FACS sorting and analysis of CD133+/CD45+ BM-HSC and peripheral blood HSC was performed, as previously described [30] and enrichment was validated by FACS analysis (to 94.0 ± 3.0%). Following antibodies were used: CD133-PE and CD133-APC (clone: MB9-3G8; Miltenyi Biotec) and CD45-FITC (clone: 30-F11; BD Bioscience).

Administration HSC

Wild-type mice received either FACS-purified CD133+/CD45+ HSCs (3 × 105 cells in 200 μl PBS) or PBS (200 μl) as control i.v. 24 h after 70% hepatectomy. Animals were sacrificed 3 d after 70% hepatectomy.

Immunohistochemistry

Liver tissue was fixed with zinc and formalin 10% and paraffin-embedded. Immunohistochemistry for Ki67 (Sigma) was performed, as described previously [29].

Statistics

Statistical analysis and graphing were performed using SPSS 12 and SigmaPlot 10. All results are expressed as mean ± SEM. Statistical significance was determined by Student's t-test, Welch test and anova and significance was defined as *P < 0.05, **P < 0.01, ***P < 0.001.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Reduced liver function depending on the extent of hepatic resection

To investigate whether the extent of liver resection has an impact on liver function, patients were evaluated by liver synthesis function tests following small (<20% TLV) and large (>30% TLV) hepatic resections. Patients with a hepatic resection greater than 30% demonstrated impaired liver function as evident by elevated bilirubin (Fig. 1A) and decreased serum albumin levels (Fig. 1B). Interestingly, no difference was detected for transaminases within the two groups (Figure S1A and B). Furthermore, a significant impairment of coagulation as an indirect sign of decreased liver synthesis was observed in patients with extended liver resections (>30%) as measured by higher INR levels (Fig. 1C), less fibrinogen (Fig. 1D), decreased Quick, and longer PTT (Figure S1C and D). No difference was detected in hepatic blood loss determined by haemoglobin and haematocrit (Figure S1E and F). Patients with larger resections beyond 30% of TLV also demonstrated decreased inflammation as evidenced by decreased fibrinogen and C-reactive protein (CRP) levels (Fig. 1D and E). Increased CRP levels, up to post-operative day (POD) 2, were observed in the small resection group, compared with large resection patients, which may reflect the greater functional liver reserve and capacity to produce this acute phase protein after surgery.

image

Figure 1. Impaired liver synthesis and function in response to large liver resections. Patients with large liver resections [>30% of total liver volume (TLV)] demonstrate decreased post-operative hepatic function as measured by serum (A) bilirubin, (B) albumin, (C) international normalized ratio (INR) and (D) fibrinogen when compared with patient group with small resections (<20% of TLV). (E) Note that the large resected group shows decreased CRP serum levels as an inability to produce acute phase proteins. (F) The indocyanine green (ICG) metabolism assay demonstrates a significant reduction in the plasma disappearing rate (PDR) up to POD 10 after extended liver resections. POD, post-operative day; *P < 0.05, **P < 0.01, ***P < 0.001.

Download figure to PowerPoint

The metabolism of ICG, a test of liver function reserve, convincingly demonstrated impaired liver function up to day 10 in the large resection group, with a massive reduction in the plasma disappearing rate (PDR) from day 1 (Fig. 1F) and an increased remaining ICG level after 15 min of ICG application (R15) (Figure S2). These ICG metabolism levels did not change in the small resection group.

Taken together, our data show an impaired hepatic function and metabolism following large liver resections.

Leucocyte and HSC mobilization in the early course of human and murine hepatic resection

WBC levels in the large resection group were superior to the small resection group, pre-operatively and on day 10 (Fig. 2A). However, mean levels were still in normal range in both groups pre-operatively and just above the normal range in the large resection group on day 10. To further investigate whether extended liver resections result in increased HSC mobilization, we looked at HSC expression in our two groups. For CD133+/CD45+ HSC, we found a significant elevation following liver surgery in the large resection group from POD 2 up to POD 90 (Fig. 2B). CD34+/CD45+ cells demonstrated similar kinetics and differences among the two groups with increased HSC mobilization after large hepatic resection, although to a lesser extend (Fig. 2C).

image

Figure 2. Increased peripheral haematopoietic stem cells (HSC) mobilization in response to large liver resections. (A) White blood cells (leucocytes) were measured from POD 0–90 in small [<20% of total liver volume (TLV)] and large (>30% of TLV) resected patients. Increased human HSC expression was detected in the large resection group for (B) CD133+/CD45+ HSC as well as (C) CD34+/CD45+ HSC as measured by fluorescence-activated cell sorting (FACS) analysis. Increased murine (D) peripheral and (E) BM stem cells were measured by FACS analysis in mice after 70% hepatectomy compared with sham-operated mice. (F) Correlation of CD133+ HSC in the blood on POD 1 with reconstituted liver volume on POD 21. POD, post-operative day; *P < 0.05, **P < 0.01.

Download figure to PowerPoint

Following 70% hepatectomy in mice, the murine kinetics were comparable to the human large resection group with increased levels of peripheral mobilized HSC, with a maximum on day 3 (Fig. 2D). Sham-operated mice did not show any CD133+/CD45+ cell mobilization into the blood. Kinetics of HSC in the BM corresponded to the peripheral numbers with a maximum on day 2 following 70% hepatectomy (Fig. 2E). When we analysed the regain of resected human liver tissue on POD 21 in the >30% resected group by CT-based volumetry, our data clearly demonstrate a strong correlation between the blood level of CD133+/CD45+ HSC and liver regain (Fig. 2F).

Our results show that the increased mobilization of HSC was directly associated with the extent of hepatectomy and suggest a role in augmenting liver regeneration following hepatic resection.

Immediate increased cytokine expression in large liver resections

As recent evidence suggested that cytokines are expressed to aid the homing of HSC following liver injury [18, 19, 31-33], we wanted to further investigate the molecular mechanism that mediates progenitor cell mobilization by analysing chemokine and growth factor expression in the course of liver resection. HGF was significantly elevated in the large resection group right after the end of the resection phase, up to 24 h post-resection, compared with the small resection group (Fig. 3A). SDF-1 demonstrated a similar expression, with significant increased levels in large resected patients up to 6 h post-resection (Fig. 3B). IGF-1 was expressed significantly higher at 3 h post-resection in the large resection group when compared with patients with resections <20% (Fig. 3C). SCF was not different among the two groups over the entire observation period (Figure S3A). In our murine model of 70% liver resection, hepatic cytokine mRNA expression was also significantly increased three- to four-fold for HGF (Fig. 3D), SDF-1 (Fig. 3E), and IGF (Fig. 3F) within the first 24 h. In contrast to our results observed in experiments performed in human peripheral blood, murine SCF mRNA expression in liver showed a moderate increase in the initial 24 h post-hepatectomy (Figure S3B).

image

Figure 3. Increased chemotactant expression in response to extended hepatic resection. (A) Human hepatic growth factor (HGF), (B) stroma-derived factor-1 (SDF-1) and (C) Insulin-like growth factor (IGF-1) serum expression were measured by enzyme-linked immunosorbent assay (ELISA) in small [<20% of total liver volume (TLV)] and large (>30% of TLV) resected groups up to 150 h. Murine mRNA expression was measured for (D) HGF, (E) SDF-1 and (F) IGF after 70% partial hepatectomy by qRT-PCR. Values are expressed as fold increase. *P < 0.05, **P < 0.01, ***P < 0.001.

Download figure to PowerPoint

Taken together, our data demonstrate increased cytokine expression in response to large liver resection as a homing effect to mobilize peripheral stem cells to promote early liver regeneration.

HGF/c-Met- and SDF-1/CXCR4-dependent mobilization of human CD133+/CD45+ HSC following liver resection

A linear positive correlation was observed between HGF serum levels within the first 6 h and corresponding CD133+/CD45+ and CD34+/CD45+ HSC mobilization on day 1 and 2 respectively (Fig. 4 A–D, Figure S4A and B). To further explore whether the observed cytokine expression positively affects peripheral stem cell mobilization, the chemotactants SDF-1 and HGF were tested for their potential for stem cell mobilization. Transwell Boyden chamber chemotaxis assays revealed specific in vitro migration towards recombinant HGF and SDF-1 gradients respectively (Fig. 4E). SDF-1 chemotaxis was significantly inhibited by the specific receptor CXCR4 antagonist AMD (Fig. 4F). Pre-incubation of CD133+/CD45+ HSC with c-Met targeting RAB prior to chemotaxis completely blocked mobilizing effects of HGF to buffer-control levels (Fig. 4F).

image

Figure 4. Chemotactant-dependent mobilization of haematopoietic stem cells (HSC) after large liver resections. Serum hepatic growth factor (HGF) expression was measured 0 and 6 h after resection and was correlated with (A,B) CD133+/CD45+ and (C,D) CD34+/CD45+ HSCs on POD 2. (E) Chemotaxis assays were performed in Transwell Boyden Chambers after pre-incubation with CD133+/CD45+ HSC for 24 h (5 μM). (F) Migration rates of CD133+/CD45+ HSC to stroma-derived factor-1 (SDF-1) (125 ng/ml), HGF (150 ng/ml) or after pre-incubation with selective inhibitors AMD (5 μg/ml) and RAB (5-10 μM). POD, post-operative day; h, hours. *P < 0.05.

Download figure to PowerPoint

Taken together, these data suggest that CD133+/CD45+ HSC are specifically mobilized by increased chemotactant release as a result of hepatic injury.

Extent of resected liver volume and HSC correlate with AFP expression

As AFP is a known fetal oncogene, hepatocellular cancer marker, but also a marker of liver progenitor cell activation, we therefore analysed AFP levels in our two resection groups. Patients with pre-operatively elevated AFP levels (mainly because of hepatocellular carcinoma) were excluded from AFP data evaluations. AFP levels were elevated in the early phase after human resection being superior in the large resection group from day 2 compared with the small resection group. However, statistical significance was achieved after POD 4 (Fig. 5A). The levels of mobilized CD133+/CD45+ HSC on POD 4 derived from all patients irrespective from the resection extend positively correlated with the subsequent AFP levels on POD 6 (Fig. 5B). These findings were further supported by increased hepatic AFP expression in mice following 70% partial hepatectomy compared with sham-operated mice on day 2.

image

Figure 5. Increased AFP expression in large liver resections followed by hepatocyte proliferation in response to haematopoietic stem cells (HSC)-triggered liver regeneration. (A) Increased serum AFP was measured in large [>30% of total liver volume (TLV)] resected patients by enzyme-linked immunosorbent assay (ELISA). (B) Correlation of CD133+ HSC in the blood on POD 4 with serum AFP on POD 6. (C) Increased murine AFP mRNA expression was detected in mice after 70% hepatectomy compared with sham-operated controls as measured by qRT-PCR. (D) Hepatocyte proliferation was measured by Ki67 staining in mice after 70% hepatectomy followed by fluorescence-activated cell sorting (FACS)-purified CD133+/CD45+ HSC application 24 h post-resection compared with control mice treated with 70% hepatectomy and PBS alone. (E)% of Ki67-positive hepatocyte nuclei were counted in each HPF on POD 3 after 70% hepatectomy in mice. POD, post-operative day; HPF, high power field; *P < 0.05.

Download figure to PowerPoint

In summary, our results suggest that AFP expression is strongly dependent on the extent of liver resection, as large liver resections lead to increased peripheral HSC mobilization followed by heightened AFP expression.

CD133+/CD45+ HSC administration promotes murine hepatic proliferation subsequent to extended liver resection

To confirm the positive effect of peripheral stem cell mobilization on hepatic liver regeneration, we analysed hepatic proliferation with and without CD133+/CD45+ HSC administration after 70% PH in mice.

On day 3 after PH, Ki67 expression in the remnant liver was significantly increased following treatment with CD133+/CD45+ HSC, compared with control mice treated with PBS following PH (Fig. 5D,E). This proliferative effect was only seen in the early phase after surgery up to day 3 and was no longer detectable on day 5 (data not shown). These results provide further evidence that peripherally mobilized HSC positively support hepatic regeneration through enhanced hepatocyte proliferation.

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

In this study, we provide further evidence that cytokines and chemotactants are important mediators for adequate HSC mobilization to promote liver regeneration after large liver resections. We demonstrate a resected volume-dependent expression of HGF, SDF-1 and IGF-1 within 24 h post-resection, followed by CD133+ HSC mobilization from day 2, predominantly following large hepatectomies. Early HGF expression correlated directly with subsequent HSC mobilization 1–2 d later. HGF and SDF-1 mobilization of CD133+ HSC was demonstrated to be dependent on their specific receptors c-Myc and CXCR4 respectively. Extensive hepatectomy with pronounced peripheral CD133+ HSC mobilization led to a marked AFP expression, a marker of hepatic progenitor cell activation [28], which may be a component of the liver regeneration response seen in this setting.

To date, the contribution of BMSCs to liver regeneration remains controversial. Some authors have reported that stem cells reconstitute the regenerating liver by transdifferentiation into primary hepatocytes [10-12, 15, 34-36]; however, this has been challenged in subsequent papers [37]. Others advocate that extrahepatic BMSC perform cell fusion [17, 38] or function as external molecular regulators for successful liver restoration [35, 39, 40]. The contribution of HSC to liver repair is generally related to the presence and severity of liver injury. However, the exact mechanisms and cellular processes are still unclear although BMSC seem to have the potential to enter different organs from the circulation to induce organ repair. We confirmed our observation that extrahepatic CD133+ HSC mobilization is augmented in response to large liver resections to induce adequate liver regeneration [5]. These findings correlate to previous reports of in vitro differentiation of human CD133+/CD45+/CD14+ HSC into hepatic lineage in response to hepatic tissue loss after hepatectomy [24]. A significant increase in circulating CD133+ cells in peripheral blood was observed 12 h after surgery [24]. In our study, we have analysed CD133+ HSC expression during a time course up to 90 d which provides novel evidence of CD133+ stem cell mobilization in a liver volume loss-dependent manner reaching a maximum in peripheral human blood on POD 2.

Interestingly, our data reveal that increased HSC mobilization post-resection was noticeable in diseased as well as in healthy livers. This is in contrast with previous findings reporting pronounced stem cell mobilization only in large hepatic resections with pre-existing chronic liver disease. The major limitation of that report, however ,is the missing volumetric or anatomical definition of ‘large’ and ‘small’ resections [25]. In our study, we have performed CT scan volumetry to measure the exact volume of resected liver tissue. In contrast to previous studies characterizing the resected extent utilizing anatomical aspects of resection, our technique of CT scan volumetry is more precise and exact.

Another intriguing finding in our study was that liver injury induces the expression of signalling mediators such as SCF, HGF and SDF-1 that may facilitate the homing and recruitment of CD133+ HSC to the damaged liver [4, 18, 19, 31-33, 41, 42].

It was previously demonstrated that SDF-1 and HGF play an important mechanistic role for stem cell migration to the liver upon chemical liver injury [4]. With our study, we have extended those findings in a different model of partial liver resection. Similar to previous findings [4], the increase in HGF and SDF-1 as stress-induced signals in response to hepatectomy leads to migration and recruitment of HSC to the injured liver. Increased serum levels of HGF have previously been observed after right hepatectomy for living-related donation in healthy individuals [22] and are known to be associated with homing of CD34+ stem cells [4]. Our findings of significant increase in HGF within the first 24 h following resection further supports our previous report of increased HGF expression [20]. Similar to HGF, both SDF-1 and IGF-1, were also expressed within the liver in the first 3–6 h post-resection. The chemokine SDF-1 (CXCL12) is a known chemoattractant of HSC in mice [4, 43] and human [44]. It has been previously reported that BM homing of CD34+/CD38- stem cells in immune-deficient NOD/SCID mice is dependent on SDF-1/CXCR4 interactions [45, 46]. Furthermore, the injection of human SDF-1 into mouse liver further enhanced hepatic migration of human stem cells [4]. It has also previously been reported that SDF-1 is up-regulated during oval/progenitor cell proliferation acting as an activator of local stem cells in the injured liver [18].

We have previously reported higher blood levels of SCF in extended hepatectomies compared to smaller resections [20]. Surprisingly, SCF levels were not significantly elevated up to POD 5 in our present study, possibly because of a more exact definition of resected liver volume levels. SCF expression has been reported to be decreased in the liver, but was significantly increased in the serum in a murine model of PH [47]. According to our data, we believe that SCF has low mobilizing capacity, but may play an important role for local homing within the liver.

AFP, a marker of oval/progenitor cells, can be used for hepatic lineage tracing [28]. Previous results have reported an AFP increase between 1 and 4 d after PH in mice [28]. Our results demonstrated a maximum AFP expression on day 2 followed by proliferation on POD 3. As shown by immunohistochemical staining, AFP is most frequently expressed around portal and periportal areas as a typical area of atypical ductal proliferation and oval cells in the canal of Hering [28] . Although during the first 4 d, cytokine secretion as well as HSC mobilization occurs in a time-dependent signalling cascade, AFP expression was significantly induced on POD 6 as a sign for dedifferentiation into hepatocytes. In mice, we observed an initial peak of AFP mRNA expression after 6 h followed by a second peak after 48 h, which correlates with the time point of maximal stem cell mobilization. AFP was induced 1 d prior to hepatocyte proliferation as shown by Ki67 immunostaining.

The peripheral mobilization of CD133+ HSC, correlated with levels of liver regeneration after hepatectomy, and an AFP/progenitor rise, a direct signalling pathway between these two cellular compartments needs to be further explored.

We have previously successfully applied human autologous CD133+ HSC in a clinical scenario by intraportal administration following portal venous embolization of right liver segments to expand left lateral hepatic segments prior to extended liver resection [5-7]. Although the exact mechanisms by which extrahepatic stem cells and their progenitor cells promote liver regeneration are not fully elucidated [38, 48, 49], our observations suggest that the chemotractants HGF and SDF-1 play an important role in HSC mobilization in the setting of acute liver injury to promote and accelerate liver regeneration. The liver proliferative effect of therapeutic stem cell applications in the clinical and preclinical scenarios of ischaemia or hepatic volume loss-triggered local regeneration may in part be because of local accumulation of administered HSC prior to physiologically chemotactant-driven mobilization and homing.

This provides a key mechanistic distinction to deepen our understanding of the proposed model (Fig. 6) in which cytokines like HGF, SCF, SDF-1 or IGF-1 are released in response to acute liver injury within hours to induce HSC mobilization and homing into the liver after 48 h. Furthermore, HSC may induce AFP expression as a sign of progenitor cell activation which may form part of the livers regenerative response.

image

Figure 6. Working model describing the role of cytokine-modulated haematopoietic stem cells (HSC) mobilization following liver resection. In response to hepatic resection as an acute liver injury stimulus, cytokines and chemotactants hepatic growth factor (HGF), stroma-derived factor-1 (SDF-1), SCF and insulin-like growth factor (IGF-1) are early expressed to induce peripheral and bone marrow mobilization of CD133+/CD45+ and CD34+/CD45+ stem cells to the liver to promote stemness and progenitor cell activation regulated by AFP. These processes will lead to hepatocyte proliferation to induce liver regeneration and repair.

Download figure to PowerPoint

In light of the data presented here, this study may lead to new strategies for regeneration, protection and treatment of various liver diseases. However, further studies are needed to explore the exact molecular mechanisms of cytokine-induced HSC homing into the liver and resultant regeneration.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Financial support: This work was supported by grants from the YAEL Foundation and the German Research Foundation (DFG NL 2509/2-1) to N.L. M.S. acknowledges grant support from German Research Foundation (DFG SCHM 2661/1-1 and 2661/1-2), German Federal Ministry of Education and Research (BMBF, PtJ-Bio 0315883) and from the 2011 Thomas E. Starzl MD Postdoctoral Fellowship Award by the American Liver Foundation (ALF). J.S.E was supported by the German Research Foundation (SCHU 1126/4-1).

Conflict of interest: The authors do not have any disclosures to report.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
liv12195-sup-0001-FigureS1.tifimage/tif37667KFig. S1 Decreased liver function in response to large liver resections.
liv12195-sup-0002-FigureS2.tifimage/tif12259KFig. S2 Decreased indocyanine metabolism in large liver resections.
liv12195-sup-0003-FigureS3.tifimage/tif7821KFig. S3 Human and murine SCF-1 expression after liver resection.
liv12195-sup-0004-FigureS4.tifimage/tif12261KFig. S4 HSC mobilization is positively correlated with early HGF chemokine expression.
liv12195-sup-0005-Legends.docWord document25K 

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.